DOE Looks to AI to Advance Next Generation of Quantum

The Energy Department’s new Quantum Genesis initiative aims to bring artificial intelligence and quantum computing together to accelerate scientific discovery, a combination national security researchers say could help overcome some of the biggest barriers to quantum computing.

“In my mind, the Genesis Mission first and foremost will help us get to performant quantum computers. You’re always going to have a classical computer controlling a quantum computer,” Sandia National Lab Computing Research Center Director Jennifer Gaudioso told GovCIO Media & Research. “With the high-performance computing and AI tools that Genesis is bringing, we will have more effective controls to integrate AI and quantum.”

In the context of today’s supercomputers, Gaudioso said quantum computing will become another tool in the broader ecosystem.

“I really think we’re headed to a heterogeneous computing future,” Gaudioso said. “Some problems will be best solved on a data flow accelerator with an AI backbone. … A quantum computer will do something that nothing else can do for us.”

AI’s Impact on Quantum

The new initiative stems from the department’s broader Genesis Mission that aims to integrate the agency’s 17 national laboratories and supercomputers into a single unified AI platform. Quantum Genesis aims to develop and deploy the world’s first fault-tolerant quantum computing capability by 2028.

Gaudioso said researchers are only beginning to understand how AI and quantum reinforce each other.

“This work of thinking about how AI and quantum can be used together is just getting started. We’re at the point where it seems real, and that wasn’t the case in the past five or 10 years,” said Gaudioso.

Why Quantum Matters

Gaudioso discussed how she sees quantum computing solving problems beyond today’s supercomputers in areas such as materials science, drug discovery and national security.

“Atoms and molecules are inherently quantum mechanical, and those are the first problems where we believe we will see quantum advantage. They’re super hard. We can only approximate them on classical computers, and we believe those are the areas where we’re going to see real advantage emerge from quantum computing,” she said.

As a nuclear weapons laboratory, Sandia could enhance its ability to model materials used in nuclear deterrence programs by developing more accurate simulations.

“We can do much better simulations if we can understand those materials from a quantum perspective with a quantum computer,” she said. “We also expect quantum computers to be really good at complex scheduling, optimization, resource allocation, and there’s obviously lots of optimization and resource allocation from a national security space where that would be really helpful.”

Progress Hinges on Error Correction

The need for a fault-tolerant computer hinges on the fact that quantum computers require a large number of qubits. The more qubits in a system, the more powerful and noisier the system. This increase in noise leads to errors in calculations.

“It’s not just about more qubits anymore,” Gaudioso said. “It’s also about better control.”

She said breakthroughs over the past several years have increased the number of operations quantum computers can perform before errors occur.

“The number of cycles you can run, the longer the computer program you can run on a quantum computer, is growing and is rapidly approaching the useful range,” she said. “That work in the last couple of years has really changed the timelines for quantum computing.”

Gaudioso said reaching that milestone will require more than advances in hardware. It will depend on collaboration across DOE’s laboratories, universities and industry to mature quantum technologies and integrate them into existing infrastructure.

She said Sandia’s role is to make emerging quantum technologies “more useful, reliable, secure and mission relevant.”

“The labs have a lot of deep knowledge, but industry has lots of advantages too. It’s going to require us to work together as a whole ecosystem to really get there,” she said.